Coming to terms with Seattle's fault

Evolving standards for seismic design leave room for interpretation - and confus

The old axiom that "the only thing constant about life is change" could have been said about the evolution of seismic design practices in Seattle.

The process is evolutionary because with the passage of time, we've come to better understand the earthquake source mechanisms in the region. But there are some very real dilemmas in developing strategies to protect the public. As understanding of our geologic and tectonic environment evolves, engineers are faced with the task of incorporating this research into codes for the design of buildings and bridges.

It's not a smooth process, and owners, building officials and consultants are on a bumpy ride.

Shaking, breaking and taking control

Seismic design hasn't had long to evolve. As a practice in the United States, it can be traced back to the 1930s. Prior to that, earthquake design had not gained widespread acceptance within the engineering community.

While earthquakes such as the 1906 San Francisco event had a devastating impact upon society, there were no codified procedures to address the hazard. It took the 1933 Long Beach earthquake to initiate legislation in California for seismic design provisions in new buildings. For the first time the design profession went beyond the static gravity loads in the design of structures, applying a fraction of the load as a horizontal force to represent the effects of earthquake ground shaking.

Then geologists and seismologists went further with their first attempts to map the severity of seismic hazards in the country. The Seismic Zone Map is the result, with the land area divided according to four levels of seismic activity.

Because California has been recognized as the hotbed of seismic activity in the U.S., coastal areas in California were given the highest seismic ranking (Zone 4) and other areas were designated as having a lessor hazard. Within this scheme, Seattle falls within zone 3, where "aseismic" areas, such as Texas, fall within zone 0. Early seismic zone maps were delineated according to historic seismic activity. The maps were changed as more information became available on past activity and faulting potential. Over the past 50 years, seismic design coefficients in Seattle have been about 75 percent of the values used for design in the coastal areas of California.

What we know and when we knew it

Significant changes in earthquake design in the Northwest were set into motion with the discoveries of Brian Atwater, a researcher with the United States Geological Survey, who hypothesized in the early 80s that land form movements observed along the Coast of Washington were related to large (magnitude 8+) earthquakes that typically occurred at intervals of 500 years.

While the conclusions of Atwater drew debate within some quarters of the scientific community, over time his findings about subduction zone earthquakes came to be widely embraced. Then the engineering community, through committees of the Structural Engineers Association of Washington (SEAW), set about incorporating these findings into the building codes. Essentially, the findings of Atwater resulted in the coastal areas of Washington (and Oregon) being upgraded from a seismic Zone 2 to a Zone 3. This process, from the inception of an idea to design codification, took nearly 10 years.

Then came the Seattle Fault, and with it another evolutionary change in local seismic design practices. The Seattle Fault is an east-west trending structure that runs between Bainbridge and Issaquah.

While the fault trace can not be readily seen at the ground surface, the potential existence of the fault was inferred from gravity and magnetic surveys dating back to the 1960s and 1970s. In fact, the Seattle Fault represents one of the largest observed in the United States. However, in the 1970s it was thought that the fault was very old and not capable of producing damaging earthquakes.

In the 1980s, further geologic studies were made of the fault and it was hypothesized that the fault may have moved about 3,000 years ago and that there was some 3,000 feet of vertical offset between the north side of the fault and the south side. It wasn't until the mid 1990's that more extensive geological investigations concluded that the fault moved 1,100 years ago.

This conclusion was based upon the dating of features that were triggered by the occurrence of the earthquake. These features included large landslides in Lake Washington, Tsunami sand deposits on Whidbey Island, raised beaches on Bainbridge and lowered beaches at West Point. In total, it appeared that over 20 feet of vertical ground movement accompanied this earthquake. While this movement did not rupture the ground surface in Seattle, it did result in some surface rupture on Bainbridge.

While the scientific community has accepted the historic occurrence of a large (magnitude 7) earthquake on the Seattle Fault and the timing of the last one can be inferred, the geologic information is less clear on typical repeat times for such earthquakes. Based on available data, it can be reasonably concluded that such events may occur at intervals of 5,000 years or longer.

This is a rare event, and one that is not normally considered in the seismic design of most structures. Specifically, the current Uniform Building Code is based upon a seismic event that has a recurrence interval of 500 years. Within this framework, ground motion contributions from events on the Seattle Fault would not be a controlling factor in seismic design because of the low probability of occurrence of these motions.

A potential earthquake on the Seattle Fault presents some interesting dilemmas for design professionals, owners and building officials. First, if a large earthquake were to occur on the Seattle Fault, it would likely result in ground motions twice as great as current design standards account. This raises the question of the level of acceptable risk for seismic design, and whether or not Seattle should adopt the standard of national practice in the current building codes or adopt a more conservative posture. The economic implications of a more conservative standard may be quite significant, considering the rarity of such an earthquake.

The design picture is further complicated by the fact that new structures are designed using different codes than the guidelines that are typically used for the seismic retrofitting of existing structures. While new structures are designed using the UBC, guidelines developed for the Federal Emergency Management Agency are typically used for the seismic retrofit of existing buildings and there are substantial differences between the two codes. Specifically, the FEMA guidelines are based on more recent geologic and tectonic data and design procedures.

The FEMA guidelines also implicitly consider the potential effects from rare earthquakes (2,500 year recurrence interval) and the guidelines have provisions for earthquake performance criteria, allowing owners to make intelligent choices on the level of service they would like their building to attain following a damaging earthquake.

The net result of the FEMA guidelines would be to place Seattle midway between Zones 3 and 4 in the current UBC. This effectively results in a condition where a building designed to meet today's standards (UBC) is seismically deficient tomorrow using the more scientifically up-to-date FEMA guidelines. This apparent discrepancy will be resolved in upcoming versions of the International Building Code (IBC 2000).

Codes and client counseling

In the meantime, owners may be reluctant to accomplish seismic retrofits because it appears that the engineering profession is not unified in approaches to the code. This potential reaction is counter productive in an environment where the design professionals and building officials are trying to encourage retrofitting to reduce losses in future earthquakes.

Faced with the above situations, the best strategy among the design professionals is one of education. Points to be stressed in discussions with owners include:

• Seattle is in earthquake country.

• Seismic building codes are changing documents that have evolved as our understanding of the geology and structure of the region has increased.

• The rates at which building codes change to adopt new information is directly related to the size of the group that is involved with the approval process.

• Discrepancies can and do exist between codes for the seismic design of new structures and the retrofitting of existing structures.

• Owner education is important to understand that buildings designed in accordance with today's codes are to protect life safety and are not necessarily guaranteed 100 percent functionality following a large earthquake.

• New building codes (FEMA and IBC 2000) will address performance issues where buildings may be designed to provide an owner specified level of functionality following a severe event.

• Older buildings, particularly unreinforced masonry structures, are clearly at risk of damage from future earthquakes and should be retrofitted to improve their seismic performance.

• While the IBC 2000 will help to clarify standards for us all, science will never rest. You can rest assured that there will be more changes in standards for seismic performance of structures in the Northwest.

---

djc home | top | special issues index